Creating Synthetic IrisCodes to Feed Biometrics Experiments
|
|
- Kelley Boyd
- 6 years ago
- Views:
Transcription
1 Creating Synthetic IrisCodes to Feed Biometrics Experiments Hugo Proença and João C. Neves Department of Computer Science IT - Instituto de Telecomunicações University of Beira Interior, Portugal {hugomcp,jcneves}@di.ubi.pt Abstract The collection of iris data suitable to be used in experiments is difficult, mainly due to two factors: ) the time spent by volunteers in the acquisition process; and ) security / privacy concerns of volunteers. Even though there are methods to create images of artificial irises, there is no method exclusively focused in the synthesis of the iris biometric signatures (IrisCodes). In experiments related with some phases of the biometric recognition process (e.g., indexing / retrieval), a large number of signatures is required for proper evaluation, which, in case of real data, is extremely hard to obtain. Hence, this paper describes a stochastic method to synthesize IrisCodes, based on the notion of data correlation. These artificial signatures can be used to feed experiments on iris recognition, namely on the iris matching, indexing and retrieval phases. We experimentally confirmed that both the genuine and impostor distributions obtained on the artificial data closely resemble the values obtained in data sets of real irises. Finally, another interesting feature is that the method is easily parametrized to mimic IrisCodes extracted from data of varying levels of quality, i.e., ranging from data acquired in high controlled to unconstrained environments. I. INTRODUCTION Among multiple traits, the iris has made rapid strides in popularity due to the remarkable effectiveness of the deployed recognition systems [] and to other interesting features: ) its texture has a randotypic chaotic appearance possible to acquire in a contactless way; ) it has a simple shape, making easier its detection and segmentation; ) it is roughly planar, enabling to compensate for deformations caused by camera-subject misalignments; and 4) most of its discriminating information lies in the lowest and middle-low frequency components of the signal, which are the most robust to noise. The nationwide deployment of iris recognition systems is considered a success. In the last information update about the UAE system [5], over million identities were included on its watch-list, and more than 50,000 deportees were prevented from entering the Emirates. The Unique Identification Authority of India [6] is deploying the system at the largest scale, with more than 00 million persons enrolled and adding about one million new identities per day, performing 6e 4 daily crosscomparisons to search for duplicate identities. To support research efforts, various iris image data sets are freely available (e.g., the CASIA [8], ICE [], WVU [4], BATH [7], MMU [], Olomuc [4] and UBIRIS []). However, at this time, these sets contain less than 0 4 identities, making it hard to objectively assess the effectiveness of algorithms on large-scale scenarios. As a response, several attempts to create artificial iris images were done, which images acceptably resemble the appearance of real data. In this paper we are particularly interested in providing data for the signatures matching and indexing / retrieval phases. We describe a stochastic method to obtain a large number of synthetic binary IrisCodes. The requirement of such type of method is evident, as generating a large number of artificial images is computationally expensive and unfeasible for practical scenarios. Also, the generation of binary signatures that closely resemble the extracted from real data is not straightforward, being important to account for the following factors: Impostors dissimilarity. The bit-by-bit comparison of signatures from different subjects should produce a large dissimilarity. The variability of these scores should be relatively small. Genuine dissimilarity. The bit-by-bit comparison of signatures from the same subject should produce a smaller dissimilarity than for the impostors. Also, the variability of these values should be significantly higher than in the case of impostors. The remainder of this paper is organized as follows: Section II summarizes the most relevant methods to synthesize iris data. Section III provides a description of the proposed method. Section IV presents and discusses the experiments. Finally, the conclusions are given in Section V. II. RELATED WORK As above stated, several methods were published to create artificial images of the iris that can be used for algorithm evaluation. However, the issue is their computational cost, which is specially concerning in case that large data sets (e.g., for over 0 9 subjects) are required. This section summarizes the most relevant methods published in this scope. Lefohn et al. [9] proposed a method to create and render realistic looking irises by adding one layer at a time to the model and rendering an intermediate result, allowing incremental definition of the iris texture, using single layers taken from their standard library of textures. This method is useful in applications ranging from entertainment to ocular prosthetics. Cui et al. [] proposed an iris synthesis method based on the analysis of principal components (PCA). They used an iris recognition algorithm based on PCA that operates on real images and allows to extract global feature vectors. These
2 ,, ρ s,, ρ a ρ i ρ g General Template ˆΩ 0 Occlusions Subject Template ˆΩ i Biometric Signatures Ω i Fig.. Cohesive perspective of the parameters evolved in the synthesis of iris signatures. The different ρ values signal the correlation parameters. The left column represents two IrisCodes from subject (each one with components extracted at two scales) and the right column illustrates an excerpt of an iriscode of subject (C (i) j,k denotes a the jth code at the k th scale, from the i th person). vectors were further used in image reconstruction. Iris samples that belong to the same class are constructed through letting the coefficients lie in the same sphere centered at a sample iris image in a high-dimensional space. To simulate different classes, they searched in a limited high-dimensional space. Also, authors concluded that super-resolution methods enhance the quality of the resulting images. Theoretical analysis and experimental results showed that the synthetic data mimics the traditional within-class and inter-class distances of real iris data. Shah et al. [5] proposed a technique to create digital versions of iris images used to evaluate the performance of iris recognition algorithms. Their scheme was divided into two phases: ) at first, a Markov Random Field model generated a background texture that represents the global iris appearance; ) next, a variety of iris features, radial and concentric furrows, collarette and crypts, were embedded in the texture field. Experiments with iris recognition algorithms validated the potential of this scheme. Zuo et al. [9] proposed a model and anatomy-based method for synthesizing iris images, having as purpose provide to the academia and industry a large data set to test iris recognition algorithms. This work also concerned about the bias that might be introduced by using synthetic data, having performed a comparison between the results observed for real and synthetic iris images. The comparison was quantified at three different levels: ) global layout, ) features of fine iris textures, and ) recognition performance, including performance extrapolation capabilities. In most cases, the results confirm their expectation of a strong similarity between real and synthetic iris data generated using their model-based approach. Wei et al. [8] proposed an iris synthesis method and claimed to establish an effective paradigm to synthesize large iris databases, with the purpose to overcome the problems of data collection. Patch-based sampling was firstly employed to create prototypes, from where a number of intra-class samples were derived from each prototype. Experiments showed that the synthetic irises preserve the major properties of real ones and bear controllable statistics, making them suitable for algorithm evaluation. III. PROPOSED METHOD According to the most acknowledged iris recognition algorithm [], an IrisCode results from the convolution between the normalised iris image and a bank of Gabor filters at different scales. Then, the signal of the resulting complex coefficients determines the binary components of the signature. In agreement to that algorithm, the method proposed in this paper generates a set of binary values grouped in different scales, using the notion of correlation. For comprehensibility, let C (i) j,k denote the jth code at the k th scale, from the i th person and c(x, y, s) denote a bit of an IrisCode at position (x, y) and scale s. Figure gives an overall perspective of the parameters evolved in the synthesis of IrisCodes. The left column shows two IrisCodes of subject, extracted at two scales. The right column gives an excerpt of the code of another subject ( ). As in real signatures, each code has n = 048 bits, with dimensions n r n c at different scales n s. Hence, four correlation parameters are used in the synthesis process: ρ a dictates the strength of the linear correlation between bits that are spatially adjacent in the biometric signature. ρ s corresponds to the strength of the linear correlation between bits extracted from the same position of the iris at different scales. ρ g is the strength of the linear correlation between the corresponding bits of different signatures of subject. Finally, ρ i corresponds to the strength of the linear correlation between bits of the same position and scale of signatures extracted from different subjects. The process is divided into three main phases: ) a general template is created, which determines the subjects templates. This general template depends of the ρ a parameter; ) next, a template is created for each virtual subject. In this case, ρ i dictates the dissimilarity between the templates of subjects; ) a set of sample IrisCodes is created for each subject, considering the ρ g parameter to control how much different will be these samples per subject; and 4) occlusions in the irises are simulated, which correspond to regions of the IrisCodes
3 where bits are purely random. Formally, let u be a random value drew from a uniform distribution U U(0, ). u is quantized into binary value, maintaining similar probabilities for 0 s and s: u q = {, if u 0.5 0, if u > 0.5 () Let ρ. be a correlation value, (either ρ a, ρ i, ρ g or ρ s ). Every bit of code c at position (x, y) is generated in top-left to bottom-right order in the following manner: c(x, y) = (H ( t r 0 r ) H(( + erf( tr ) ρ. ) ) ) u q () being t r 0 is the total number of 0 bits in a neighbourhood of radius r, erf is the sigmoid error function, the exclusive OR logical operation and H the Heaviside function, given by: H(x) = { 0, if x 0, if x > 0 The top-left bit of the general template of the data set is purely random. Then, all the bits in the subjects template are generated according to (), using ρ a as correlation parameter and r =. Next, the first scale of the templates for each subject is generated, using the ρ i value and obtaining t r 0 from the generic template. For all subsequent scales, ρ s controls the correlation and t r 0 is taken from the anterior scale. In a third step, the samples per subject are created, according to the ρ g value and taking t r 0 from the subject template at the corresponding scale. In order to simulate different quality acquisition environments, a quality parameter ξ [0, ] weights the values of ρ g, i.e., ρ g = ξρ g. The final step simulates the regions of the iris that are occluded (e.g., due to eyelids or eyelashes). Two semi-circles of radii r and r are draw and placed in the bottom part of the IrisCode, which are known to be the regions that are most frequently occluded in the normalized images. Next, considering that bits extracted from eyelids or eyelashes do not possess any discriminating ability, bits inside these circles are generated in a purely random way (), disregarding all the correlation ρ values. Figure illustrates two occluded regions in IrisCodes, where the image at the top represents an heavily occluded image, in opposition to the bottom image that is almost noise-free. Table III summarises the parameters evolved in the above described synthesis process, giving the range of values allowed for each one. Additionally, the bottom rows of that Table give examples of the parameters used to simulate environments of ideal conditions, and heavily unconstrained environments. These values were used to generate the data sets of Env. A and Env. D and are given for guidance of readers. Examples of the IrisCodes generated are shown in Figure, illustrating the effect of the ρ a parameter. Here, large values () Fig.. Illustration of the bits in IrisCodes that are extracted from regions of the iris occluded by eyelids or eyelashes. These regions do not possess any discriminating ability between the genuine and impostors comparisons and - as such - the corresponding bits are draw in a purely random way (). Parameter Range Description ρ s [0,] Scale correlation. Controls the probability that bits extracted from the same positions of the iris at different scales have similar value. ρ a [0,] Spatial correlation. Controls the probability that bits extracted from adjacent positions of the iris have similar values. ρ g [0,] Genuine correlation. Controls the probability that bits extracted from images of a given subject have similar values. ρ i [0,] Impostors correlation. Controls the probability that bits extracted from images of different subjects have similar values. ξ [0,] Corresponds directly to the quality of the data generated. 0 corresponds to data of poorest quality and simulates signatures extracted from high quality data. Optimal Environment ρ s=0.5, ρ a=0., ρ g=0.7, ρ i=0.05, ξ= Unconst. Environment ρ s=0.5, ρ a=0., ρ g=0., ρ i=0.07, ξ=0 TABLE I. SUMMARY OF THE PARAMETERS EVOLVED IN THE PROPOSED METHOD FOR THE SYNTHESIS OF IrisCodes. increase the correlation between adjacent bits (upper rows), whereas small values decrease this dependency and turn (for ρ a = 0) the values of each bit independent of its neighborhood. The upper row of Figure 4 illustrates the ρ s parameter. Here, two-scale signatures from subjects and are shown. The bottom row gives the effect of ρ g by showing two additional signatures of subject. The bottommost table gives the pairwise distances between IrisCodes, confirming that all requirements about codes dissimilarity were faithfully modeled. IV. EXPERIMENTS Figure 5 shows two histograms of the genuine (dashed lines) and impostor (continuous lines) matching scores, with respect to the ξ parameter. In all these plots, results regard 50,000 IrisCodes from 0,000 simulated different subjects. Previous studies shown that the conditions in the acquisition environment have a strong effect in the genuine comparisons, which was also confirmed in our observations. The topmost figure gives the distributions for an environment of relatively good quality (Env. A). Then, for the remaining environments, quality decreases and, in the case of Env. D, there is a significant overlap between both distributions, as it happens in uncontrolled scenarios. Additionally, the synthetic IrisCodes were validated in terms of the performance attained by three state-of-the-art
4 ρ a = 0.9 Env. A ρ a = 0.75 (0.4, 0.05) ρ a = 0.5 ρ a = 0. Env. B (0., 0.08) ρ a = 0.0 Fig.. Effect of the parameter ρ a, that determines the spatial correlation between adjacent bits. Larger values augment the probability that neighbour codes have similar values, whereas the zero value turns the value of a bit independent of its spatial location. Env. C (0.7, 0.08) ρ s = 0.05 ρ s = 0. Env. D ρ g = 0.75 ρ g = 0.5 (0.4, 0.07) Fig. 4. Images at the top row illustrate the effect of the ρ s value. Images at the bottom illustrate the effect of ρ g. and are signatures from different subjects. The bottommost table gives the pairwise Hamming distances between and. indexing / retrieval strategies, comparing the results to the ones reported by authors in their experiments. The selected methods are due to: ) Gadde et al. [6], which analyzed the distribution of intensities and selected patterns with low coefficients of variation (CVs) as indexing pivots. For each probe represented in the polar domain, a radial division of n- bands was performed and indexed using the radial band of Fig. 5. Illustration of the separation between genuine (dashed lines) and impostor (continuous lines) comparisons, for different levels of quality. the highest density of CV patterns. ) Hao et al. [7] used the spatial spread of the most reliable bits, they propose an indexing technique based on the notion of multi-collisions. In the retrieval process, a minimum of k collisions between the probe and gallery samples is required to identify a potential match. Finally, ) Mukherjee and Ross [0] approached the problem from two different perspectives, by analyzing the iris texture and the IrisCode. The best results in the latter case were attained when each code was split into fixed-size blocks. First-order statistics for each block were used as the primary indexing value. A k-means strategy was used to divide the feature space into different classes. For comprehensibility, a single numeric score was used to assess levels of performance, in terms of the relation between
5 Method Real Env. A Env. B Env. C Env. D Gadde et al. [6] Hao et al. [7] Mukherjee and Ross [0] TABLE II. RESULTS OBTAINED BY THREE STATE-OF-THE-ART IRIS INDEXING / RETRIEVAL METHODS ON SIGNATURES EXTRACTED FROM REAL IRISES (COLUMN Real) AND USING THE SYNTHETIC IrisCodes GENERATED BY THE PROPOSED METHOD (COLUMNS Env. A-D). the hit and penetration rates, as suggested by Mukherjee and Ross [0]: τ = h( p), (4) being h and p the hit and penetration rates. Table IV compares the results announced by authors in their experiments (Column Real) to the results obtained for synthetic IrisCodes, according to the method proposed in this paper. For contextualization, four different environments are shown (columns Env. A to Env. D), corresponding to the histograms of Figure 5. For both the methods of Gadde et al. and Mukherjee and Ross, the results observed for synthetic data were poorer than those reported by authors, enabling to conclude about an extremely high quality level of the images used in their experiments. Also, we noted that both indexing methods are extremely sensitive to slight changes in the distributions of genuine / importer scores. Specifically, they significantly increase their effectiveness when the genuine distribution is positively skewed, which does not happens in the generated data sets. In the case of the method of Hao et al., results obtained in the synthetic IrisCodes were close to the reported by authors, specially in the case of environment A (highlighted in bold), in which the genuine / impostor distributions closely resemble the results given by authors. This fact was positively regarded as a strong indicator of the quality of the synthetic codes. V. CONCLUSIONS Aiming to resemble the IrisCodes that result from the most acknowledged iris recognition algorithm (Daugman s []), this paper described a stochastic method to generate synthetic IrisCodes, based on the notion of linear correlation. This method can be used to create an extremely large number of iris signatures, used to evaluate / validate different phases of the iris recognition process (e.g., iris matching, indexing / retrieval algorithms). When performing an all-against-all comparison between the generated codes, we confirmed that the resulting genuine and impostor matching scores faithfully resemble the corresponding distributions observed for real iris data. Also, an additional empirical validation was carried out by comparing the results obtained by three state-of-the-art indexing / retrieval techniques on real and artificial IrisCodes.The easy parameterization of the proposed method should be highlighted, to resemble the conditions in acquisition environments of varying quality. REFERENCES [] J. Cui, Y. Wang, J. Huang, T. Tan, and Z. Sun. An iris image synthesis method based on PCA and super-resolution. Proceedings of the IEEE, 94():97 95, 006. [] J. Daugman. Probing the uniqueness and randomness of Iriscodes: Results from 00 billion iris pair comparisons. Proceedings of the 7th International Conference on Pattern Recognition, 4:47 474, 004. [] J. Daugman. How Iris Recognition Works. IEEE Transactions on Circuits and Systems for Video Technology, 4(): 0, 004. [4] M. Dobes and L. Machala. [online], [5] Emirates ID Head Office. Iris scan prevents entry of 50,000 deportees: Saif Bin Zayed. assessed on June 0. [6] R. Gadde, D. Adjeroh, and A. Ross. Indexing iris images using the burrows-wheeler transform. Proceedings of the IEEE International Workshop on Information Forensics and Security (WIFS), pages 6, 00. [7] F. Hao, J. Daugman, and P. Zielinski. A fast search algorithm for a large fuzzy database. IEEE Transactions on Information Forensics and Security, ():0, 008. [8] Institute of Automation, Chinese Academy of Sciences. [online] http: // [9] A. Lefohn, B. Budge, P. Shirley, R. Caruso, and E. Reinhard. An ocularist s approach to human iris synthesis. Computer Graphics and Applications, (6):70 75, 00. [0] R. Mukherjee and A. Ross. Indexing iris images. Proceedings of the 9th International Conference on Pattern Recognition, pages 4, 008. [] Multimedia University. [online] [] National Institute of Standards and Technology. [online] gov/ice/. [] H. Proença, S. Filipe, R. Santos, J. Oliveira, and L. A. Alexandre. The ubiris.v: A database of visible wavelength iris images captured on-themove and at-a-distance. IEEE Transactions on Pattern Analysis and Machine Intelligence, (8):50 56, 00. [4] A. Ross, S. Crihalmeanu, L. Hornak, and S. Schuckers. A centralized web-enabled multimodal biometric database. Proceedings of the 004 Biometric Consortium Conference (BCC), 004. [5] S. Shah and A. Ross. Generating synthetic irises by feature agglomeration. Proceedings of the 006 IEEE International Conference on Image Processing, pages 7 0, 006. [6] Unique Identification Authority of India. [online], about-uidai.html, accessed on June, 0. [7] University of Bath. [online], sipg/. [8] Z. Wei, T. Tan, and Z. Sun. Synthesis of large realistic iris databases using patch-based sampling. Proceedings of the 9th International Conference on Pattern Recognition, pages 4, 008. [9] J. Zuo, N. A. Schmid, and X. Chen. On generation and analysis of synthetic iris images. IEEE Transactions on Information Forensics and Security, ():77 90, 007.
Graph Matching Iris Image Blocks with Local Binary Pattern
Graph Matching Iris Image Blocs with Local Binary Pattern Zhenan Sun, Tieniu Tan, and Xianchao Qiu Center for Biometrics and Security Research, National Laboratory of Pattern Recognition, Institute of
More informationComputationally Efficient Serial Combination of Rotation-invariant and Rotation Compensating Iris Recognition Algorithms
Computationally Efficient Serial Combination of Rotation-invariant and Rotation Compensating Iris Recognition Algorithms Andreas Uhl Department of Computer Sciences University of Salzburg, Austria uhl@cosy.sbg.ac.at
More informationIris Recognition: Measuring Feature s Quality for the Feature Selection in Unconstrained Image Capture Environments
CIHSPS 2006 - IEEE INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE FOR HOMELAND SECURITY AND PERSONAL SAFETY, ALEXANDRIA,VA, USA, 16-17 O Iris Recognition: Measuring Feature s Quality for the Feature
More informationOn Generation and Analysis of Synthetic Iris Images
1 On Generation and Analysis of Synthetic Iris Images Jinyu Zuo, Student Member, IEEE, Natalia A. Schmid, Member, IEEE, and Xiaohan Chen, Student Member, IEEE Department of Computer Science and Electrical
More informationCritique: Efficient Iris Recognition by Characterizing Key Local Variations
Critique: Efficient Iris Recognition by Characterizing Key Local Variations Authors: L. Ma, T. Tan, Y. Wang, D. Zhang Published: IEEE Transactions on Image Processing, Vol. 13, No. 6 Critique By: Christopher
More informationA Method for the Identification of Inaccuracies in Pupil Segmentation
A Method for the Identification of Inaccuracies in Pupil Segmentation Hugo Proença and Luís A. Alexandre Dep. Informatics, IT - Networks and Multimedia Group Universidade da Beira Interior, Covilhã, Portugal
More informationIRIS recognition II. Eduard Bakštein,
IRIS recognition II. Eduard Bakštein, edurard.bakstein@fel.cvut.cz 22.10.2013 acknowledgement: Andrzej Drygajlo, EPFL Switzerland Iris recognition process Input: image of the eye Iris Segmentation Projection
More informationL Iride. Docente: Michele Nappi. biplab.unisa.it
L Iride Docente: Michele Nappi mnappi@unisa.it biplab.unisa.it 089-963334 AGENDA 1. Biometric Overview 2. IRIS and Daugman 3. NICE I: Iris Segmentation (Detection) 4.NICE II: Iris Coding and Matching (Recognition)
More informationEnhanced Iris Recognition System an Integrated Approach to Person Identification
Enhanced Iris Recognition an Integrated Approach to Person Identification Gaganpreet Kaur Research Scholar, GNDEC, Ludhiana. Akshay Girdhar Associate Professor, GNDEC. Ludhiana. Manvjeet Kaur Lecturer,
More informationIris Recognition for Eyelash Detection Using Gabor Filter
Iris Recognition for Eyelash Detection Using Gabor Filter Rupesh Mude 1, Meenakshi R Patel 2 Computer Science and Engineering Rungta College of Engineering and Technology, Bhilai Abstract :- Iris recognition
More informationA Coding Scheme for Indexing Multimodal Biometric Databases
A Coding Scheme for Indexing Multimodal Biometric Databases Aglika Gyaourova West Virginia University Morgantown WV 26506, USA agyaouro@csee.wvu.edu Arun Ross West Virginia University Morgantown WV 26506,
More informationAn Efficient Iris Recognition Using Correlation Method
, pp. 31-40 An Efficient Iris Recognition Using Correlation Method S.S. Kulkarni 1, G.H. Pandey 2, A.S.Pethkar 3, V.K. Soni 4, &P.Rathod 5 Department of Electronics and Telecommunication Engineering, Thakur
More informationIris Recognition: Analyzing the Distribution of the Iriscodes Concordant Bits
Iris Recognition: Analyzing the Distribution of the Iriscodes Concordant Bits Gil Santos and Hugo Proença Department of Computer Science Instituto de Telecomunicações - Networks and Multimedia Group University
More informationA biometric iris recognition system based on principal components analysis, genetic algorithms and cosine-distance
Safety and Security Engineering VI 203 A biometric iris recognition system based on principal components analysis, genetic algorithms and cosine-distance V. Nosso 1, F. Garzia 1,2 & R. Cusani 1 1 Department
More informationSIC-Gen: ASynthetic Iris-Code Generator
A. Brömme, C. Busch, A. Dantcheva,C.Rathgeb and A. Uhl (Eds.): BIOSIG 2017, Lecture Notes in Informatics (LNI), Gesellschaft fur Informatik, Bonn 2017 61 SIC-Gen: ASynthetic Iris-Code Generator Pawel Drozdowski
More informationAlgorithms for Recognition of Low Quality Iris Images. Li Peng Xie University of Ottawa
Algorithms for Recognition of Low Quality Iris Images Li Peng Xie University of Ottawa Overview Iris Recognition Eyelash detection Accurate circular localization Covariance feature with LDA Fourier magnitude
More informationNoisy iris recognition: a comparison of classifiers and feature extractors
Noisy iris recognition: a comparison of classifiers and feature extractors Vinícius M. de Almeida Federal University of Ouro Preto (UFOP) Department of Computer Science (DECOM) viniciusmdea@gmail.com Vinícius
More informationIRIS SEGMENTATION OF NON-IDEAL IMAGES
IRIS SEGMENTATION OF NON-IDEAL IMAGES William S. Weld St. Lawrence University Computer Science Department Canton, NY 13617 Xiaojun Qi, Ph.D Utah State University Computer Science Department Logan, UT 84322
More informationInternational Journal of Advance Engineering and Research Development. Iris Recognition and Automated Eye Tracking
International Journal of Advance Engineering and Research Development Scientific Journal of Impact Factor (SJIF): 4.72 Special Issue SIEICON-2017,April -2017 e-issn : 2348-4470 p-issn : 2348-6406 Iris
More informationChapter 5. Effective Segmentation Technique for Personal Authentication on Noisy Iris Images
110 Chapter 5 Effective Segmentation Technique for Personal Authentication on Noisy Iris Images Automated authentication is a prominent goal in computer vision for personal identification. The demand of
More informationA New Encoding of Iris Images Employing Eight Quantization Levels
A New Encoding of Iris Images Employing Eight Quantization Levels Oktay Koçand Arban Uka Computer Engineering Department, Epoka University, Tirana, Albania Email: {okoc12, auka}@epoka.edu.al different
More informationTutorial 8. Jun Xu, Teaching Asistant March 30, COMP4134 Biometrics Authentication
Tutorial 8 Jun Xu, Teaching Asistant csjunxu@comp.polyu.edu.hk COMP4134 Biometrics Authentication March 30, 2017 Table of Contents Problems Problem 1: Answer The Questions Problem 2: Daugman s Method Problem
More informationCOMPUTATIONALLY EFFICIENT SERIAL COMBINATION OF ROTATION-INVARIANT AND ROTATION COMPENSATING IRIS RECOGNITION ALGORITHMS
COMPUTATIONALLY EFFICIENT SERIAL COMBINATION OF ROTATION-INVARIANT AND ROTATION COMPENSATING IRIS RECOGNITION ALGORITHMS Mario Konrad, Herbert Stögner School of Communication Engineering for IT, Carinthia
More informationPeriocular Biometrics: When Iris Recognition Fails
Periocular Biometrics: When Iris Recognition Fails Samarth Bharadwaj, Himanshu S. Bhatt, Mayank Vatsa and Richa Singh Abstract The performance of iris recognition is affected if iris is captured at a distance.
More informationIRIS Recognition System Based On DCT - Matrix Coefficient Lokesh Sharma 1
Volume 2, Issue 10, October 2014 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online at: www.ijarcsms.com
More informationAn Application of ARX Stochastic Models to Iris Recognition
An Application of ARX Stochastic Models to Iris Recognition Luis E. Garza Castañón 1, Saúl Montes de Oca 2, and Rubén Morales-Menéndez 1 1 Department of Mechatronics and Automation, ITESM Monterrey Campus,
More informationComparing Binary Iris Biometric Templates based on Counting Bloom Filters
Christian Rathgeb, Christoph Busch, Comparing Binary Iris Biometric Templates based on Counting Bloom Filters, In Proceedings of the 18th Iberoamerican Congress on Pattern Recognition (CIARP 13), LNCS
More informationShifting Score Fusion: On Exploiting Shifting Variation in Iris Recognition
Preprocessing c 211 ACM This is the author s version of the work It is posted here by permission of ACM for your personal use Not for redistribution The definitive version was published in: C Rathgeb,
More informationAre Iris Crypts Useful in Identity Recognition?
Are Iris Crypts Useful in Identity Recognition? Feng Shen Patrick J. Flynn Dept. of Computer Science and Engineering, University of Notre Dame 254 Fitzpatrick Hall, Notre Dame, IN 46556 fshen1@nd.edu flynn@nd.edu
More informationBiorthogonal wavelets based Iris Recognition
Biorthogonal wavelets based Iris Recognition Aditya Abhyankar a, Lawrence Hornak b and Stephanie Schuckers a,b a Department of Electrical and Computer Engineering, Clarkson University, Potsdam, NY 13676,
More informationHuman Identification from at-a-distance Face Images using Sparse Representation of Local Iris Features
Human Identification from at-a-distance Face Images using Sparse Representation of Local Iris Features Ajay Kumar, Tak-Shing Chan, Chun-Wei Tan Department of Computing, The Hong Kong Polytechnic University,
More informationIMPROVED FACE RECOGNITION USING ICP TECHNIQUES INCAMERA SURVEILLANCE SYSTEMS. Kirthiga, M.E-Communication system, PREC, Thanjavur
IMPROVED FACE RECOGNITION USING ICP TECHNIQUES INCAMERA SURVEILLANCE SYSTEMS Kirthiga, M.E-Communication system, PREC, Thanjavur R.Kannan,Assistant professor,prec Abstract: Face Recognition is important
More informationFast and Efficient Automated Iris Segmentation by Region Growing
Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 2, Issue. 6, June 2013, pg.325
More informationSignature Recognition by Pixel Variance Analysis Using Multiple Morphological Dilations
Signature Recognition by Pixel Variance Analysis Using Multiple Morphological Dilations H B Kekre 1, Department of Computer Engineering, V A Bharadi 2, Department of Electronics and Telecommunication**
More informationToward Noncooperative Iris Recognition: A Classification Approach Using Multiple Signatures
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 29, NO. 4, APRIL 2007 607 Toward Noncooperative Iris Recognition: A Classification Approach Using Multiple Signatures Hugo Proença and
More informationIris Segmentation and Recognition System
Iris Segmentation and Recognition System M. Karpaga Kani, Dr.T. Arumuga MariaDevi Abstract-- The richness and apparent stability of the iris texture make it a robust bio-metric trait for personal authentication.
More informationClustering CS 550: Machine Learning
Clustering CS 550: Machine Learning This slide set mainly uses the slides given in the following links: http://www-users.cs.umn.edu/~kumar/dmbook/ch8.pdf http://www-users.cs.umn.edu/~kumar/dmbook/dmslides/chap8_basic_cluster_analysis.pdf
More informationHaresh D. Chande #, Zankhana H. Shah *
Illumination Invariant Face Recognition System Haresh D. Chande #, Zankhana H. Shah * # Computer Engineering Department, Birla Vishvakarma Mahavidyalaya, Gujarat Technological University, India * Information
More informationEfficient Iris Identification with Improved Segmentation Techniques
Efficient Iris Identification with Improved Segmentation Techniques Abhishek Verma and Chengjun Liu Department of Computer Science New Jersey Institute of Technology Newark, NJ 07102, USA {av56, chengjun.liu}@njit.edu
More informationColor Local Texture Features Based Face Recognition
Color Local Texture Features Based Face Recognition Priyanka V. Bankar Department of Electronics and Communication Engineering SKN Sinhgad College of Engineering, Korti, Pandharpur, Maharashtra, India
More informationDEFORMABLE MATCHING OF HAND SHAPES FOR USER VERIFICATION. Ani1 K. Jain and Nicolae Duta
DEFORMABLE MATCHING OF HAND SHAPES FOR USER VERIFICATION Ani1 K. Jain and Nicolae Duta Department of Computer Science and Engineering Michigan State University, East Lansing, MI 48824-1026, USA E-mail:
More informationSSRG International Journal of Electronics and Communication Engineering (SSRG-IJECE) Volume 3 Issue 6 June 2016
Iris Recognition using Four Level HAAR Wavelet Transform: A Literature review Anjali Soni 1, Prashant Jain 2 M.E. Scholar, Dept. of Electronics and Telecommunication Engineering, Jabalpur Engineering College,
More informationAn Efficient Iris Recognition System using Phase Based Technique
ISSN No: 2454-9614 An Efficient Iris Recognition System using Phase Based Technique T.Manickam, A.Sharmila, A.K.Sowmithra Department Of Electronics and Communications Engineering, Nandha Engineering College,
More informationThe Elimination Eyelash Iris Recognition Based on Local Median Frequency Gabor Filters
Journal of Information Hiding and Multimedia Signal Processing c 2015 ISSN 2073-4212 Ubiquitous International Volume 6, Number 3, May 2015 The Elimination Eyelash Iris Recognition Based on Local Median
More informationInternational Journal of Emerging Technologies in Computational and Applied Sciences (IJETCAS)
International Association of Scientific Innovation and Research (IASIR) (An Association Unifying the Sciences, Engineering, and Applied Research) ISSN (Print): 2279-0047 ISSN (Online): 2279-0055 International
More informationLearning Based Enhancement Model of Iris
Learning Based Enhancement Model of Iris Junzhou Huang, Li Ma, Tieniu Tan, Yunhong Wang National Laboratory of Pattern Recognition, Institute of Automation Chinese Academy of Sciences, P.O. Box 2728, Beijing,
More informationA Fast and Accurate Eyelids and Eyelashes Detection Approach for Iris Segmentation
A Fast and Accurate Eyelids and Eyelashes Detection Approach for Iris Segmentation Walid Aydi, Lotfi Kamoun, Nouri Masmoudi Department of Electrical National Engineering School of Sfax Sfax University
More informationIncremental Iris Recognition: A Single-algorithm Serial Fusion Strategy to Optimize Time Complexity
C Rathgeb, A Uhl, and P Wild Iris Recognition: A Single-algorithm Serial Fusion Strategy to Optimize Time Complexity In Proceedings of the IEEE 4th International Conference on Biometrics: Theory, Applications,
More informationPartial Face Recognition
Partial Face Recognition Shengcai Liao NLPR, CASIA April 29, 2015 Background Cooperated face recognition People are asked to stand in front of a camera with good illumination conditions Border pass, access
More informationALGORITHM FOR BIOMETRIC DETECTION APPLICATION TO IRIS
ALGORITHM FOR BIOMETRIC DETECTION APPLICATION TO IRIS Amulya Varshney 1, Dr. Asha Rani 2, Prof Vijander Singh 3 1 PG Scholar, Instrumentation and Control Engineering Division NSIT Sec-3, Dwarka, New Delhi,
More informationGurmeet Kaur 1, Parikshit 2, Dr. Chander Kant 3 1 M.tech Scholar, Assistant Professor 2, 3
Volume 8 Issue 2 March 2017 - Sept 2017 pp. 72-80 available online at www.csjournals.com A Novel Approach to Improve the Biometric Security using Liveness Detection Gurmeet Kaur 1, Parikshit 2, Dr. Chander
More informationImage Enhancement Techniques for Fingerprint Identification
March 2013 1 Image Enhancement Techniques for Fingerprint Identification Pankaj Deshmukh, Siraj Pathan, Riyaz Pathan Abstract The aim of this paper is to propose a new method in fingerprint enhancement
More informationAn efficient face recognition algorithm based on multi-kernel regularization learning
Acta Technica 61, No. 4A/2016, 75 84 c 2017 Institute of Thermomechanics CAS, v.v.i. An efficient face recognition algorithm based on multi-kernel regularization learning Bi Rongrong 1 Abstract. A novel
More informationFeature-level Fusion for Effective Palmprint Authentication
Feature-level Fusion for Effective Palmprint Authentication Adams Wai-Kin Kong 1, 2 and David Zhang 1 1 Biometric Research Center, Department of Computing The Hong Kong Polytechnic University, Kowloon,
More informationPersonal Identification with Human Iris Recognition Based on EMD
Personal Identification with Human Iris Recognition Based on EMD Abstract: Shubhangi D C 1, Nageshwari Anakal 2 1( Professor, Department of studies in Computer Science & Engineering,) 2( P.G.Student, Department
More informationNon-Ideal Iris Segmentation Using Graph Cuts
Workshop on Biometrics (in association with CVPR) Anchorage, Alaska, June 2008 Non-Ideal Iris Segmentation Using Graph Cuts Shrinivas J. Pundlik Damon L. Woodard Stanley T. Birchfield Electrical and Computer
More informationIRIS SEGMENTATION AND RECOGNITION FOR HUMAN IDENTIFICATION
IRIS SEGMENTATION AND RECOGNITION FOR HUMAN IDENTIFICATION Sangini Shah, Ankita Mandowara, Mitesh Patel Computer Engineering Department Silver Oak College Of Engineering and Technology, Ahmedabad Abstract:
More informationProblem definition Image acquisition Image segmentation Connected component analysis. Machine vision systems - 1
Machine vision systems Problem definition Image acquisition Image segmentation Connected component analysis Machine vision systems - 1 Problem definition Design a vision system to see a flat world Page
More informationIris Recognition using Four Level Haar Wavelet Transform
Iris Recognition using Four Level Haar Wavelet Transform Anjali Soni 1, Prashant Jain 2 M.E. Scholar, Dept. of Electronics and Telecommunication Engineering, Jabalpur Engineering College, Jabalpur, Madhya
More informationSemi-Fragile Watermarking in Biometric Systems: Template Self-Embedding
Semi-Fragile Watermarking in Biometric Systems: Template Self-Embedding Reinhard Huber 1, Herbert Stögner 1, and Andreas Uhl 1,2 1 School of CEIT, Carinthia University of Applied Sciences, Austria 2 Department
More informationCURRENT iris recognition systems claim to perform
1 Improving Iris Recognition Performance using Segmentation, Quality Enhancement, Match Score Fusion and Indexing Mayank Vatsa, Richa Singh, and Afzel Noore Abstract This paper proposes algorithms for
More informationFinal Exam Schedule. Final exam has been scheduled. 12:30 pm 3:00 pm, May 7. Location: INNOVA It will cover all the topics discussed in class
Final Exam Schedule Final exam has been scheduled 12:30 pm 3:00 pm, May 7 Location: INNOVA 1400 It will cover all the topics discussed in class One page double-sided cheat sheet is allowed A calculator
More informationSeminary Iris Segmentation. BCC448 Pattern Recognition
Seminary Iris Segmentation BCC448 Pattern Recognition Students: Filipe Eduardo Mata dos Santos Pedro Henrique Lopes Silva Paper Robust Iris Segmentation Based on Learned Boundary Detectors Authors: Haiqing
More informationChapter-2 LITERATURE REVIEW ON IRIS RECOGNITION SYTSEM
Chapter-2 LITERATURE REVIEW ON IRIS RECOGNITION SYTSEM This chapter presents a literature review of iris recognition system. The chapter is divided mainly into the six sections. Overview of prominent iris
More informationImproved Iris Segmentation Algorithm without Normalization Phase
Improved Iris Segmentation Algorithm without Normalization Phase R. P. Ramkumar #1, Dr. S. Arumugam *2 # Assistant Professor, Mahendra Institute of Technology Namakkal District, Tamilnadu, India 1 rprkvishnu@gmail.com
More informationImplementation of Reliable Open Source IRIS Recognition System
Implementation of Reliable Open Source IRIS Recognition System Dhananjay Ikhar 1, Vishwas Deshpande & Sachin Untawale 3 1&3 Dept. of Mechanical Engineering, Datta Meghe Institute of Engineering, Technology
More informationAn Adaptive Threshold LBP Algorithm for Face Recognition
An Adaptive Threshold LBP Algorithm for Face Recognition Xiaoping Jiang 1, Chuyu Guo 1,*, Hua Zhang 1, and Chenghua Li 1 1 College of Electronics and Information Engineering, Hubei Key Laboratory of Intelligent
More informationHuman Gait Recognition Using Bezier Curves
Human Gait Recognition Using Bezier Curves Pratibha Mishra Samrat Ashok Technology Institute Vidisha, (M.P.) India Shweta Ezra Dhar Polytechnic College Dhar, (M.P.) India Abstract-- Gait recognition refers
More informationComparative Study of ROI Extraction of Palmprint
251 Comparative Study of ROI Extraction of Palmprint 1 Milind E. Rane, 2 Umesh S Bhadade 1,2 SSBT COE&T, North Maharashtra University Jalgaon, India Abstract - The Palmprint region segmentation is an important
More informationFace Recognition Technology Based On Image Processing Chen Xin, Yajuan Li, Zhimin Tian
4th International Conference on Machinery, Materials and Computing Technology (ICMMCT 2016) Face Recognition Technology Based On Image Processing Chen Xin, Yajuan Li, Zhimin Tian Hebei Engineering and
More informationThe Impact of Diffuse Illumination on Iris Recognition
The Impact of Diffuse Illumination on Iris Recognition Amanda Sgroi, Kevin W. Bowyer, and Patrick J. Flynn University of Notre Dame asgroi kwb flynn @nd.edu Abstract Iris illumination typically causes
More informationFast trajectory matching using small binary images
Title Fast trajectory matching using small binary images Author(s) Zhuo, W; Schnieders, D; Wong, KKY Citation The 3rd International Conference on Multimedia Technology (ICMT 2013), Guangzhou, China, 29
More informationA Review of Image- based Rendering Techniques Nisha 1, Vijaya Goel 2 1 Department of computer science, University of Delhi, Delhi, India
A Review of Image- based Rendering Techniques Nisha 1, Vijaya Goel 2 1 Department of computer science, University of Delhi, Delhi, India Keshav Mahavidyalaya, University of Delhi, Delhi, India Abstract
More informationA multi-biometric verification system for the privacy protection of iris templates
A multi-biometric verification system for the privacy protection of iris templates S. Cimato, M. Gamassi, V. Piuri, R. Sassi and F. Scotti Dipartimento di Tecnologie dell Informazione, Università di Milano,
More informationA Survey on Feature Extraction Techniques for Palmprint Identification
International Journal Of Computational Engineering Research (ijceronline.com) Vol. 03 Issue. 12 A Survey on Feature Extraction Techniques for Palmprint Identification Sincy John 1, Kumudha Raimond 2 1
More informationBeyond Bags of Features
: for Recognizing Natural Scene Categories Matching and Modeling Seminar Instructed by Prof. Haim J. Wolfson School of Computer Science Tel Aviv University December 9 th, 2015
More informationInternational Journal of Advanced Research in Computer Science and Software Engineering
ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: Fingerprint Recognition using Robust Local Features Madhuri and
More informationA Contactless Palmprint Recognition Algorithm for Mobile Phones
A Contactless Palmprint Recognition Algorithm for Mobile Phones Shoichiro Aoyama, Koichi Ito and Takafumi Aoki Graduate School of Information Sciences, Tohoku University 6 6 05, Aramaki Aza Aoba, Sendai-shi
More informationEfficient Iris Spoof Detection via Boosted Local Binary Patterns
Efficient Iris Spoof Detection via Boosted Local Binary Patterns Zhaofeng He, Zhenan Sun, Tieniu Tan, and Zhuoshi Wei Center for Biometrics and Security Research National Laboratory of Pattern Recognition,
More informationA Novel Identification System Using Fusion of Score of Iris as a Biometrics
A Novel Identification System Using Fusion of Score of Iris as a Biometrics Raj Kumar Singh 1, Braj Bihari Soni 2 1 M. Tech Scholar, NIIST, RGTU, raj_orai@rediffmail.com, Bhopal (M.P.) India; 2 Assistant
More informationDr. Enrique Cabello Pardos July
Dr. Enrique Cabello Pardos July 20 2011 Dr. Enrique Cabello Pardos July 20 2011 ONCE UPON A TIME, AT THE LABORATORY Research Center Contract Make it possible. (as fast as possible) Use the best equipment.
More informationA NEW OBJECTIVE CRITERION FOR IRIS LOCALIZATION
The Nucleus The Nucleus, 47, No.1 (010) The Nucleus A Quarterly Scientific Journal of Pakistan Atomic Energy Commission NCLEAM, ISSN 009-5698 P a ki sta n A NEW OBJECTIVE CRITERION FOR IRIS LOCALIZATION
More informationRobust and Secure Iris Recognition
Robust and Secure Iris Recognition Vishal M. Patel University of Maryland, UMIACS pvishalm@umiacs.umd.edu IJCB 2011 Tutorial Sparse Representation and Low-Rank Representation for Biometrics Outline Iris
More informationHOUGH TRANSFORM CS 6350 C V
HOUGH TRANSFORM CS 6350 C V HOUGH TRANSFORM The problem: Given a set of points in 2-D, find if a sub-set of these points, fall on a LINE. Hough Transform One powerful global method for detecting edges
More informationFingerprint Image Enhancement Algorithm and Performance Evaluation
Fingerprint Image Enhancement Algorithm and Performance Evaluation Naja M I, Rajesh R M Tech Student, College of Engineering, Perumon, Perinad, Kerala, India Project Manager, NEST GROUP, Techno Park, TVM,
More informationTowards Securing Biometric Templates using Self Generated Dynamic Helper Data
, June 30 - July 2, 2010, London, U.K. Towards Securing s using Self Generated Dynamic Helper Data Aditya Abhyankar and Stephanie Schuckers Abstract information is irrevokable and hence cannot be compromised.
More informationA Comparative Study of Palm Print Recognition Systems
A Comparative Study of Palm Print Recognition Systems Akash Patel akash.patel@somaiya.edu Chinmayi Tidke chinmayi.t@somaiya.edu Chirag Dhamecha Mumbai India chirag.d@somaiya.edu Kavisha Shah kavisha.shah@somaiya.edu
More informationA Framework for Efficient Fingerprint Identification using a Minutiae Tree
A Framework for Efficient Fingerprint Identification using a Minutiae Tree Praveer Mansukhani February 22, 2008 Problem Statement Developing a real-time scalable minutiae-based indexing system using a
More informationSchedule for Rest of Semester
Schedule for Rest of Semester Date Lecture Topic 11/20 24 Texture 11/27 25 Review of Statistics & Linear Algebra, Eigenvectors 11/29 26 Eigenvector expansions, Pattern Recognition 12/4 27 Cameras & calibration
More informationImage Matching Using Run-Length Feature
Image Matching Using Run-Length Feature Yung-Kuan Chan and Chin-Chen Chang Department of Computer Science and Information Engineering National Chung Cheng University, Chiayi, Taiwan, 621, R.O.C. E-mail:{chan,
More informationwww.worldconferences.org Implementation of IRIS Recognition System using Phase Based Image Matching Algorithm N. MURALI KRISHNA 1, DR. P. CHANDRA SEKHAR REDDY 2 1 Assoc Prof, Dept of ECE, Dhruva Institute
More informationA Propitious Iris Pattern Recognition Using Neural Network Based FFDTD and HD Approach
International Journal of Computer Science and Telecommunications [Volume 5, Issue 12, December 2014] 13 ISSN 2047-3338 A Propitious Iris Pattern Recognition Using Neural Network Based FFDTD and HD Approach
More informationLocal Correlation-based Fingerprint Matching
Local Correlation-based Fingerprint Matching Karthik Nandakumar Department of Computer Science and Engineering Michigan State University, MI 48824, U.S.A. nandakum@cse.msu.edu Anil K. Jain Department of
More informationBIOMET: A Multimodal Biometric Authentication System for Person Identification and Verification using Fingerprint and Face Recognition
BIOMET: A Multimodal Biometric Authentication System for Person Identification and Verification using Fingerprint and Face Recognition Hiren D. Joshi Phd, Dept. of Computer Science Rollwala Computer Centre
More informationA Fast Personal Palm print Authentication based on 3D-Multi Wavelet Transformation
A Fast Personal Palm print Authentication based on 3D-Multi Wavelet Transformation * A. H. M. Al-Helali, * W. A. Mahmmoud, and * H. A. Ali * Al- Isra Private University Email: adnan_hadi@yahoo.com Abstract:
More informationLearning and Inferring Depth from Monocular Images. Jiyan Pan April 1, 2009
Learning and Inferring Depth from Monocular Images Jiyan Pan April 1, 2009 Traditional ways of inferring depth Binocular disparity Structure from motion Defocus Given a single monocular image, how to infer
More informationImproved Iris Recognition in 2D Eigen Space
Improved Iris Recognition in 2D Eigen Space Abhijit Das School of Education Technology Jadavpur University Kolkata, India Ranjan Parekh School of Education Technology Jadavpur University Kolkata India
More informationRobust Iris Recognition based on Orientation Field
Robust Iris Recognition based on Orientation Field Pravin S.Patil Department of Electronics and Communication S. S. V. P. S s B. S. Deore College of Engineering, Dhule, (M.S.), India ABSTRACT Recently,
More informationCombined Fingerprint Minutiae Template Generation
Combined Fingerprint Minutiae Template Generation Guruprakash.V 1, Arthur Vasanth.J 2 PG Scholar, Department of EEE, Kongu Engineering College, Perundurai-52 1 Assistant Professor (SRG), Department of
More informationVerifying Fingerprint Match by Local Correlation Methods
Verifying Fingerprint Match by Local Correlation Methods Jiang Li, Sergey Tulyakov and Venu Govindaraju Abstract Most fingerprint matching algorithms are based on finding correspondences between minutiae
More information